Anti-cancer screening at the National Cancer Institute
Efforts to identify compounds that might have utility in the treatment of cancer started with Cancer Chemotherapy National Service Center, created by act of Congress in 1955. A Symposium was held to mark the 50th anniversary of the program and part of that effort included assembling a series of interviews that gives an extremely valuable historical perspective not just on NCI's drug discovery efforts, but this history of anti-cancer drug discovery in general. A recording of the entire symposium is online and although at almost 5 hours it is a lot to get though, it is a fascinating look at present and past thinking on anti-cancer drug discovery. It is particularly interesting to note that there has never been a time when there was clear agreement on how to attack this very difficult problem. A timeline was also assembled (available in a Flash version and a non-Flash version) that provides the interviews and text in smaller, easier to navigate chunks.
For roughly the first thirty years of the NCI testing program, the primary screen was activity in a mouse leukemia model. There are a number of problems with this process.
It takes a significant amount of compound to run the test
The tests are much more complicated and expensive than in vitro tests
Although a number of useful drugs were identified, they tended to have very poor activity in common solid tumors (breast, lung, colon). This suggested that it was very important to have a screen that included a greater variety of tumor types.
Technological progress had reached a point in the mid 1980s where it was feasible to develop a replacement for the mouse models. In 1990 the human tumor cell line assay went into production as the primary screen. The screen consists of measuring the ability of compounds to inhibit the growth of 60 human tumor cell lines grown in culture. The 60 cell lines cover nine disease types. Experimental details can be found here.
As mentioned above, there has never been anything remotely close to unanimous agreement on how best to find anti-cancer drugs. One of the basic debates is between empirical screening versus targeted screening. Empirical screening uses an assay that, at least at some level, uses an intact system that captures the property you are interested in. In the present case, this means looking for compounds that might inhibit the growth of tumors in patients by looking for compounds that inhibit the growth of human tumor cells in culture. Targeted screening uses biological information to pick a particular biochemical activity, say inhibition of a particular enzyme, and then creates a screen that looks very specifically for that biochemical activity. Each strategy has strengths and weaknesses. If a compound is inactive in an empirical screen it can be very difficult and time consuming to figure out exactly why it failed; was it because it couldn't get past the cell membrane?, because it couldn't reach the targets it needed to reach inside the cell?, because it was metabolized to inactive forms? Without answers to these kinds of questions it is essentially impossible to plan what to do next with any confidence. Even if a compound is active in an empirical assay, there are challenges. It can be very difficult and time consuming to figure out what the compound is doing at a biochemical level to lead to the activity you are seeing. This is especially a problem in anti cancer drugs as it is increasingly clear that cancer, even within a particular diagnosis, is a very heterogeneous disease on a molecular level. Matching a patient's molecular pathology with drug(s) that can benefit him will certainly require a good understanding of the molecular basis for the drug's action. This is, at least in principle, not a problem for a targeted discovery effort. You have designed the discovery effort to produce compounds that work in a certain way. Of course one problem is that there are always surprises and just because a compound possesses a certain activity, it doesn't necessarily mean that that is the activity that is responsible for the biological activity in complex systems. It is also not uncommon to spend considerable effort designing compounds that have maximal effect in the cell free system, only to find out that the final compounds suffer from any number of problems that prevent it from showing any activity in more complex systems.
In the mid to late 1990s a substantial part of the cancer research community thought that empirical screening was not really relevant anymore and that the NCI should discontinue the cell line screen. A review was organized and the final report recognized a real value of the cell screen and recommended that it be continued, although with a change in emphasis. The work of Ken Paull was a major factor in the recognition of the review panel that empirical screening had a useful place in cancer drug discovery. Paull developed the COMPARE alogrithm, which analyzes the pattern of activity across all the cell lines in the screen. Paull realized that compounds whose pattern of activity were similar generally also had similar molecular mechanisms of action. This showed that a panel of empirical assays combined with appropriate data analysis could capture some of the advantages of targeted screens while retaining the advantages of an empirical screen. A large part of these wiki pages and this collaboration will be focused on demonstrating how this can be useful to academic synthetic organic chemists.
Anti-cancer screening at the National Cancer Institute
Efforts to identify compounds that might have utility in the treatment of cancer started with Cancer Chemotherapy National Service Center, created by act of Congress in 1955. A Symposium was held to mark the 50th anniversary of the program and part of that effort included assembling a series of interviews that gives an extremely valuable historical perspective not just on NCI's drug discovery efforts, but this history of anti-cancer drug discovery in general. A recording of the entire symposium is online and although at almost 5 hours it is a lot to get though, it is a fascinating look at present and past thinking on anti-cancer drug discovery. It is particularly interesting to note that there has never been a time when there was clear agreement on how to attack this very difficult problem. A timeline was also assembled (available in a Flash version and a non-Flash version) that provides the interviews and text in smaller, easier to navigate chunks.
For roughly the first thirty years of the NCI testing program, the primary screen was activity in a mouse leukemia model. There are a number of problems with this process.
- It takes a significant amount of compound to run the test
- The tests are much more complicated and expensive than in vitro tests
- Although a number of useful drugs were identified, they tended to have very poor activity in common solid tumors (breast, lung, colon). This suggested that it was very important to have a screen that included a greater variety of tumor types.
Technological progress had reached a point in the mid 1980s where it was feasible to develop a replacement for the mouse models. In 1990 the human tumor cell line assay went into production as the primary screen. The screen consists of measuring the ability of compounds to inhibit the growth of 60 human tumor cell lines grown in culture. The 60 cell lines cover nine disease types. Experimental details can be found here.As mentioned above, there has never been anything remotely close to unanimous agreement on how best to find anti-cancer drugs. One of the basic debates is between empirical screening versus targeted screening. Empirical screening uses an assay that, at least at some level, uses an intact system that captures the property you are interested in. In the present case, this means looking for compounds that might inhibit the growth of tumors in patients by looking for compounds that inhibit the growth of human tumor cells in culture. Targeted screening uses biological information to pick a particular biochemical activity, say inhibition of a particular enzyme, and then creates a screen that looks very specifically for that biochemical activity. Each strategy has strengths and weaknesses. If a compound is inactive in an empirical screen it can be very difficult and time consuming to figure out exactly why it failed; was it because it couldn't get past the cell membrane?, because it couldn't reach the targets it needed to reach inside the cell?, because it was metabolized to inactive forms? Without answers to these kinds of questions it is essentially impossible to plan what to do next with any confidence. Even if a compound is active in an empirical assay, there are challenges. It can be very difficult and time consuming to figure out what the compound is doing at a biochemical level to lead to the activity you are seeing. This is especially a problem in anti cancer drugs as it is increasingly clear that cancer, even within a particular diagnosis, is a very heterogeneous disease on a molecular level. Matching a patient's molecular pathology with drug(s) that can benefit him will certainly require a good understanding of the molecular basis for the drug's action. This is, at least in principle, not a problem for a targeted discovery effort. You have designed the discovery effort to produce compounds that work in a certain way. Of course one problem is that there are always surprises and just because a compound possesses a certain activity, it doesn't necessarily mean that that is the activity that is responsible for the biological activity in complex systems. It is also not uncommon to spend considerable effort designing compounds that have maximal effect in the cell free system, only to find out that the final compounds suffer from any number of problems that prevent it from showing any activity in more complex systems.
In the mid to late 1990s a substantial part of the cancer research community thought that empirical screening was not really relevant anymore and that the NCI should discontinue the cell line screen. A review was organized and the final report recognized a real value of the cell screen and recommended that it be continued, although with a change in emphasis. The work of Ken Paull was a major factor in the recognition of the review panel that empirical screening had a useful place in cancer drug discovery. Paull developed the COMPARE alogrithm, which analyzes the pattern of activity across all the cell lines in the screen. Paull realized that compounds whose pattern of activity were similar generally also had similar molecular mechanisms of action. This showed that a panel of empirical assays combined with appropriate data analysis could capture some of the advantages of targeted screens while retaining the advantages of an empirical screen. A large part of these wiki pages and this collaboration will be focused on demonstrating how this can be useful to academic synthetic organic chemists.
Next Page Previous Page